the evil twin relationship between crypto and ML is becoming more sharply defined by the week now. Whoever unifies them conceptually and mathematically will basically rule the future.

Blogger at ribbonfarm.com, running summerofprotocols.com, host of /gloom ☀️

A market maker that intentionally eats toxic flow at relatively small size on many venues then feeds this data to an ML that uses it to discern sharp flow from soft flow on the main venue at size

Building Dyad: https://dyadstable.xyz

FID 1479

Deep learning; quantitative finance; DeFi.

you would bait sharp takers on every relevant venue by running smaller mm as a sacrificial sensor, gather data before being run over and use this to train the ml to detect this toxic flow and protect the “real” mm at size